Methods for maintaining electronic devices within a safe and desirable operating temperature range have been a topic of research since the invention of the transistor. Maintaining such a temperature range is a challenging problem that is only increasing in importance and difficulty as semiconductor technology continues to progress. State of the art microprocessors easily produce more than 40 W of thermal energy per square centimeter of the microchip surface. Power electronics can attain heat densities three times this level.
In addition to the requirement to manage such high heat intensity, there is a need to remove the thermal energy efficiently, both in terms of energy expended and space required. According to the Department of Energy, approximately 3% of electricity used in the United States is devoted to powering data centers or computer facilities. Approximately half of this electricity goes toward power conditioning and cooling. Increasing the efficiency of cooling would lead to dramatic savings in energy. More efficient cooling is also needed in transportation systems due to the rapidly increased adoption of hybrid and electric vehicles. More efficient cooling of the electronic systems in these vehicles translates into increased range and utility of the vehicles.
The majority of computer systems are currently cooled using air that is forced through a series of extended metal surfaces coupled to microchips or other electronic work pieces. However, these systems are inherently limited in terms of their performance and efficiency. Due to the very low volumetric heat capacity of air, a large volume of air flow is required to remove the heat load of even one processor. A recommended value is 5 to 10 cubic feet per minute (cfm) per 100 W of heat load. This equates to the equivalent of two air conditioning systems sized for a typical U.S. house being required to cool a rack of computers. A typical data center may have several hundred of these racks.
Furthermore, air-cooled systems are not only inefficient in themselves but also cause the electronics they cool to operate less efficiently. Because of the low thermal capacity of air, fully utilized microprocessors operate at or near the maximum rated temperature. Reducing the temperature of microprocessors can save at least 25% of the energy they consume at the same level of utilization.
Numerous liquid cooling schemes have been implemented to address some of the problems associated with air cooling. A majority rely on using water that flows through channels defined by fins, wherein the fins are indirectly coupled to a work piece via a metal base plate, a thermal paste, and a direct bond metal such as copper. This approach can be effective. However, the intervening materials between the water and the work piece induce significant thermal resistance, which reduces the efficiency of the system. In addition to the thermal resistance, the intervening materials add to the cost and time of manufacture, constitute additional points of failure, and provide possible disposal issues. Finally, the intervening materials render the system unable to efficiently deal with local hot spots on a work piece. The entire system must be designed to accommodate the maximum anticipated heat load of one or a few localized hot spots.
Further improvements have been made to liquid-cooled systems by using a coolant other than water. Dielectric coolants can come into direct contact with the electronic devices and not harm them. Use of such dielectric coolants permits eliminating a significant amount of thermal interface material from the system. However, the dielectric coolants are less efficient coolants than water. More aggressive cooling techniques are therefore required to achieve the necessary performance.
One approach with dielectric coolants includes direct spray impingement, in which atomized liquid coolant is sprayed directly on a work piece surface through air or vapor. However, spray cooling is limited by several factors. First, spray cooling requires a significant working volume to enable the atomized sprays to form. Second, atomizing the liquid requires a significant amount of pressure upstream of the atomizer to generate an appropriate pressure drop at the atomizer-air interface to enable atomized sprays to form. Maintaining this amount of pressure consumes a significant amount of energy. Third, high flow rates are required to prevent critical heat flux, wherein evaporation of coolant on the surface prevents atomized liquid from reaching the surface. In the end, it has proven difficult to design a practical, compact spray cooling system, despite the large amount of effort that has been expended to do so.
Another approach is to use direct jet impingement, wherein streams of liquid are projected through a liquid medium and impinge directly on a work piece surface. While impinging jets are known to have notable heat transfer performance, impinging jet systems have problems of scalability. To achieve high heat transfer over a large area, arrays of jets must be used. The use of arrays in conventional direct jet impingement systems, however, is problematic. Opposing surface flow of fluid from neighboring jet streams induces stagnant regions on the surface. The heat transfer performance in these stagnant regions can drop to nearly zero. Furthermore, conventional jet impingement systems use nozzles that are part of a large, flat nozzle plate. As fluid from jet streams impinging on the surface flow from the center of the plate flows outward, it can have enough momentum to completely deflect the outermost jets, preventing them from impinging on the heated surface. As a result of these factors, conventional impinging jet systems are limited in size.
Efficient, scalable methods and devices for cooling surfaces of work pieces are needed.